[Scilab-Dev] Weird issue in the Scilab core

[mottelet at dma.utc.fr]

Selon Sylvestre Ledru <sylvestre.ledru at inria.fr>:

> 
> Hi,
> 
> I am trying to port Scilab under Mac OS X and I have a weird issue on basic
> Scilab feature.
> 
> Using the function size is returning an error about the type of the input
> argument.
> 
> -->size(rand(1,1))
>                 !--error 42 
> Incompatible input argument.
> 
> Before I spend more time on this issue, did anyone had it before ?
> 
> Cheers,
> Sylvestre
> 

After rebuilding a complete gcc4.3 suite, and using it
for gcc, g++ and gfortran (instead of Apple gcc4.01 and
gfortran from gcc4.0.3 suite), Scilab exhibits exactly 
the same problem. I am not sure it is a general (OSX specific)
parser bug, since defining a macro (not with deff)

-->function y=f(x)
-->y=x^2
-->endfunction

and using it later is ok. Seems that only primitives
are subject to a systematic error 42 or 41.

Serge, do you have a hint ?

S.




 



-------------------------------------------------